In the example below, we’re going to use the deploy.py tool to create an
API Gateway instance called irc-hooky and add a /github endpoint to
this API. We’ll then deploy the IRC Hooky Lambda function and wire it all up!
Keep in mind that that the deploy.py tool was intended to be idempotent.
This means that it should not matter if you run it more than once. This also
means that these same instructions also apply to deploying updated versions
of IRC Hooky.
Note that these instructions are meant to be interchangeable so if you (for
example) feel the need to create a /bitbucket endpoint, set the
REST_ENDPOINT_NAME environment variable to bitbucket and that should be
all you need to do.
With all that, here are the base pre-requisites you’ll need in order to get going:
Save a copy of both lambda.zip and deploy.py in a local directory.
deploy.py needs a few parameters set via environment variables, so let’s do
that now:
export AWS_ACCESS_KEY_ID="<YOUR AWS ACCESS KEY>"
export AWS_SECRET_ACCESS_KEY="<YOUR AWS SECRET ACCESS KEY>"
export AWS_DEFAULT_REGION="us-east-1"
export LAMBDA_FUNCTION_NAME="irc-hooky"
export REST_ENDPOINT_NAME="github"
export IRCHOOKY_IRC_SERVER="chat.freenode.net"
export IRCHOOKY_IRC_PORT="6667"
export IRCHOOKY_IRC_CHANNEL="##testtest"
Do pay attention to the region and other parameters above and set them to whatever works best for you.
Now let’s create the virtualenv and install the dependencies we’re going to need:
rm -rf deploy-env
virtualenv deploy-env
deploy-env/bin/pip install requests
deploy-env/bin/pip install boto3
And the last thing we need to do here is wire everything up!
deploy-env/bin/python deploy.py
And you’re done! You should then see an output line that looks something like:
INFO:irchooky:IRC Hooky has been successfully deployed to: https://1234556x.execute-api.us-east-1.amazonaws.com/prod/github